Beidou and laser sensor adaptive time alignment method based on vibration characteristics

Abstract

The invention provides a Beidou and laser sensor adaptive time alignment method based on vibration characteristics, and relates to the technical field of Beidou multi-sensor data fusion deformation monitoring, and the method comprises the steps: collecting a sensing data set of a plurality of modal sensors, and obtaining Beidou positioning coordinate data and laser displacement sensing data; interpolating the initial Beidou positioning time sequence and the initial laser displacement time sequence to a public time reference to obtain a Beidou positioning time sequence and a laser displacement time sequence; hilbert transformation is carried out on the laser displacement time sequence, an instantaneous envelope is obtained, and a vibration section is identified; calculating the time offset between the Beidou positioning time sequence and the laser displacement time sequence according to a segmented cross-correlation algorithm, and calculating the confidence coefficient of each vibration segment; and carrying out time offset rationality check and progressive correction based on the time offset and the confidence coefficient to obtain global time offset so as to carry out global correction on the timestamp of the time sequence. According to the invention, the time alignment precision and reliability can be improved.

Description

Technical Field

[0001] This invention relates to the field of BeiDou multi-sensor data fusion deformation monitoring technology, and in particular to an adaptive time alignment method for BeiDou and laser sensors based on vibration characteristics. Background Technology

[0002] The BeiDou Navigation Satellite System (BDS) offers advantages such as all-weather operation, high real-time performance, and high automation in deformation monitoring. However, in bridge or high-rise building monitoring scenarios, BeiDou signal propagation is susceptible to multipath effects, leading to lower monitoring accuracy in complex environments and necessitating collaborative measurements with other sensors. In multi-sensor collaborative measurement systems, time synchronization issues arise between different sensors due to differences in hardware clocks, data transmission delays, and inconsistent sampling rates.

[0003] Chinese Patent CN114071694B discloses a method, device, and storage medium for improving timing accuracy based on BeiDou satellite signals. This scheme measures the jitter of each received second pulse signal based on a clock signal generated by a local clock. The measured jitter value is then used to calculate the time deviation between the second pulse signal and the corresponding second pulse signal on the local clock. This time deviation is then used to compensate for the jitter in each received second pulse signal, thus achieving jitter reduction for the BeiDou positioning receiver's corresponding second pulse signal. Finally, the time signal and the compensated second pulse signal are simulated and forwarded, essentially converting one signal into multiple identical signals for output. However, the aforementioned comparative scheme only directly converts time sampling to spatial sampling based on the vertical jitter during interference signal processing. The algorithm focuses more on the consistency correction of each frame of data without segmenting and constraining the overall scanning process, leading to accumulated phase calculation errors in complex vibration environments. Therefore, it is essential to provide an adaptive time alignment method for BeiDou and laser sensors based on vibration characteristics to improve time alignment accuracy and reliability. Summary of the Invention

[0004] In view of this, the present invention proposes an adaptive time alignment method for BeiDou and laser sensors based on vibration characteristics.

[0005] This invention provides an adaptive time alignment method for BeiDou and laser sensors based on vibration characteristics, the method comprising: Collect sensor datasets from multiple modal sensors and preprocess the sensor datasets to obtain BeiDou positioning coordinate data and laser displacement sensing data. The initial BeiDou positioning time series in the BeiDou positioning coordinate data and the initial laser displacement time series in the laser displacement sensing data are interpolated to a common time reference to obtain a BeiDou positioning time series and a laser displacement time series located on a unified time axis. Perform Hilbert transform on the laser displacement time series to obtain the instantaneous envelope of the analytical signal corresponding to the laser displacement time series, and identify the vibration segment in the laser displacement time series based on the instantaneous envelope; The time offset between the BeiDou positioning time series and the laser displacement time series is calculated using a segmented cross-correlation algorithm, and the confidence level corresponding to each vibration segment is calculated. Based on the time offset and confidence level of all vibration segments, a time offset rationality check and progressive correction operation are performed to obtain a global time offset. Then, the timestamps of the BeiDou positioning time series or laser displacement time series are globally corrected and output according to the global time offset.

[0006] Based on the above technical solutions, preferably, the preprocessing of the sensor dataset specifically includes: The sensor dataset is subjected to coarse outlier detection based on the solution quality identifier, wherein the solution quality identifier includes fixed solution or floating-point solution; Outlier detection is performed on each type of data in the sensor dataset using a sliding window local median absolute deviation. Data points marked as outliers in the sensor dataset are removed and gaps are filled using linear interpolation. After interpolation and imputation of the sensor dataset, the mean of each type of data in the sensor dataset is calculated and the mean is removed to output the BeiDou positioning coordinate data and the laser displacement sensing data.

[0007] Based on the above technical solutions, preferably, the acquisition of the BeiDou positioning time series and laser displacement time series located on a unified time axis specifically includes: The initial BeiDou positioning time series and the initial laser displacement time series are resampled onto a common time reference using the piecewise cubic Hermite interpolation method, and the resampled initial BeiDou positioning time series are filtered to obtain a BeiDou positioning time series and a laser displacement time series located on a unified time axis.

[0008] More preferably, the step of identifying vibration segments in the laser displacement time series based on the instantaneous envelope specifically includes: Perform Hilbert transform on the laser displacement time series to obtain the Hilbert transform components corresponding to the laser displacement time series, and construct an analytical signal based on the Hilbert transform components; Calculate the instantaneous envelope corresponding to the analytic signal, and smooth the instantaneous envelope by median filtering to obtain a smooth envelope; The vibration detection threshold is calculated based on the smooth envelope, and the time interval in which the smooth envelope continuously exceeds the vibration detection threshold and the duration exceeds the preset minimum duration is marked as a vibration segment.

[0009] More preferably, the time offset reasonableness check operation specifically includes: If the difference in time offset between any two adjacent vibration segments is greater than a preset continuity threshold, then the two adjacent vibration segments are marked as discontinuous abnormal segments. If the absolute value of the time offset of any vibration segment exceeds the preset time offset threshold, the vibration segment will be marked as an abnormal segment that is outside the reasonable range. If the confidence level corresponding to any vibration segment is less than the preset confidence threshold, then the vibration segment is marked as a low-quality abnormal segment.

[0010] More preferably, the progressive correction operation specifically includes: Select a reference vibration segment from the vibration segments with the highest confidence level, and use the time offset corresponding to the reference vibration segment as the global time offset reference. For abnormal vibration segments located at the beginning of the BeiDou positioning time series or the laser displacement time series, the time offset of the abnormal vibration segments is directly corrected to the global time offset reference. For abnormal vibration segments located at the end of the BeiDou positioning time series or the laser displacement time series, the time offset of the abnormal vibration segment is corrected to the time offset of the previous non-abnormal vibration segment. For abnormal vibration segments located in the middle of the BeiDou positioning time series or the laser displacement time series, the time offset of the abnormal vibration segment is corrected to the average of the time offset of the previous non-abnormal vibration segment and the global time offset reference.

[0011] More preferably, the step of globally correcting and outputting the timestamps of the BeiDou positioning time series or laser displacement time series based on the global time offset specifically includes: Interpolate or extrapolate the time offsets of all vibration segments on the time axis to obtain a continuous time offset curve, and perform global correction on all timestamps of the BeiDou positioning time series or the laser displacement time series based on the time offset curve.

[0012] A second aspect of this application provides a BeiDou and laser sensor adaptive time alignment system based on multi-source data fusion. The BeiDou and laser sensor adaptive time alignment system includes a data acquisition module, a data processing module, and a global correction module. The data acquisition module is used to acquire sensing datasets from multiple modal sensors and preprocess the sensing datasets to obtain BeiDou positioning coordinate data and laser displacement sensing data. The data processing module is used to interpolate the initial BeiDou positioning time series in the BeiDou positioning coordinate data and the initial laser displacement time series in the laser displacement sensing data to a common time reference to obtain a BeiDou positioning time series and a laser displacement time series located on a unified time axis. The module performs a Hilbert transform on the laser displacement time series to obtain the instantaneous envelope of the analytical signal corresponding to the laser displacement time series. Based on the instantaneous envelope, the module identifies vibration segments in the laser displacement time series. The module calculates the time offset between the BeiDou positioning time series and the laser displacement time series according to a piecewise cross-correlation algorithm and calculates the confidence level corresponding to each vibration segment. The global correction module is used to perform time offset rationality verification and progressive correction operations based on the time offset and confidence level of all vibration segments to obtain the global time offset, and to globally correct and output the timestamps of the Beidou positioning time series or laser displacement time series according to the global time offset.

[0013] A third aspect of this application provides an electronic device including a processor, a memory, a user interface, and a network interface, wherein the memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory.

[0014] In a fourth aspect, this application provides a computer-readable storage medium having a computer program stored thereon, the computer program being executed by a processor to implement the steps of a vibration-based adaptive time alignment method for BeiDou and laser sensors.

[0015] The vibration-based adaptive time alignment method for BeiDou and laser sensors provided by this invention has the following advantages over existing technologies: (1) By interpolating the BeiDou positioning time series and the laser displacement time series to a common time reference, and then combining the segmented cross-correlation and vibration feature identification, the precise time synchronization of the two types of heterogeneous sensor data can be achieved, which can significantly reduce the time deviation and improve the accuracy of multi-source data fusion. Furthermore, by obtaining the instantaneous envelope of the laser displacement signal through Hilbert transform and automatically identifying the vibration segment, the time offset estimation is only performed on the time period containing obvious dynamic features, which can effectively avoid invalid or unreliable alignment calculations in the static or low dynamic stage. At the same time, the segmented cross-correlation algorithm is used to calculate the time offset of each vibration segment separately, and a confidence index is assigned to each offset, which can effectively quantify the contribution and credibility of different time periods to the alignment result. The time offset and confidence obtained from all vibration segments are used to check the rationality of the time offset, which can eliminate or weaken the offset result that is inconsistent with the overall trend. The global time offset is obtained through progressive correction, realizing the transition from local estimation to global unified offset, reducing the impact of single local segment error on the final result, and improving the time alignment accuracy and reliability.

[0016] (2) By using coarse outlier detection based on solution quality identifiers, data segments with obviously unreliable positioning solution quality can be preferentially removed to avoid low-quality solutions causing significant interference to subsequent processing. Fine detection is performed using the absolute deviation of the local median in a sliding window, which can identify anomalous data such as sudden change points, spikes, and short-term drifts at the local scale, thus cleaning up noise more accurately. Furthermore, linear interpolation is used to smoothly fill gaps, ensuring the continuity of the time series and avoiding time alignment deviations and feature distortions caused by data breakpoints. The mean of each type of data after interpolation is calculated and mean removal is performed, so that both BeiDou positioning coordinate data and laser displacement data are represented in the form of zero mean, which is conducive to highlighting dynamic change features such as vibration. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 A flowchart illustrating the adaptive time alignment method for BeiDou and laser sensors based on vibration characteristics provided by this invention; Figure 2 A comparison diagram of the time-aligned BeiDou and laser displacement responses provided for this invention; Figure 3 A comparative diagram of the BeiDou and laser displacement time series provided by the present invention under different time alignment methods; Figure 4 Error normal distribution diagrams under three working conditions provided by this invention; Figure 5 This is a schematic diagram of the structure of the Beidou and laser sensor adaptive time alignment system provided by the present invention; Figure 6 This is a schematic diagram of the structure of the electronic device provided by the present invention.

[0019] Explanation of reference numerals in the attached figures: 1. Beidou and laser sensor adaptive time alignment system; 11. Data acquisition module; 12. Data processing module; 13. Global correction module; 2. Electronic equipment; 21. Processor; 22. Communication bus; 23. User interface; 24. Network interface; 25. Memory. Detailed Implementation

[0020] The technical solutions of the present invention will be clearly and completely described below with reference to the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0021] This invention discloses an adaptive time alignment method for BeiDou and laser sensors based on vibration characteristics, with reference to... Figure 1 The steps of this method include S1 to S5.

[0022] Step S1: Collect sensor datasets from multiple modal sensors and preprocess the sensor datasets to obtain BeiDou positioning coordinate data and laser displacement sensing data.

[0023] In this step, the acquired data includes BeiDou positioning coordinate data and laser displacement sensing data. For the BeiDou positioning coordinate data, coarse outlier detection is performed based on the solution quality identifier. A sliding window local median absolute deviation detection is then applied to the two sequences. Detected outliers are marked and removed, and small gaps are filled using linear interpolation. After outlier removal, the mean of the two sequences is calculated, and mean-reduction processing is performed to ensure that subsequent alignment focuses only on the dynamic characteristics of the signal, unaffected by absolute values.

[0024] This step also includes steps S11 to S13.

[0025] Step S11: Perform coarse outlier detection on the sensor dataset based on the solution quality identifier, wherein the solution quality identifier includes fixed solution or floating-point solution.

[0026] In this step, outlier detection is performed on the sensor dataset based on the solution quality identifier. Specifically, for each epoch of BeiDou positioning data in the sensor dataset, its corresponding solution quality identifier is read. This identifier can be of type such as fixed solution or floating-point solution. When the solution quality identifier indicates no solution, solution failure, or accuracy not meeting preset requirements, the BeiDou positioning data point corresponding to that epoch is directly marked as an outlier. When the solution quality identifier indicates a floating-point solution, quality indicators such as coordinate jump variables, number of satellites, or positioning covariance for that epoch can be optionally considered. If the quality indicators exceed preset thresholds, the data point corresponding to that epoch is marked as an outlier. When the solution quality identifier indicates a fixed solution and all quality indicators meet preset requirements, the data point corresponding to that epoch is marked as a normal data point. Through this coarse detection based on the solution quality identifier, obviously unreliable BeiDou positioning data is eliminated, providing more reliable input for subsequent fine-grained detection of local median absolute deviation.

[0027] Step S12: Perform outlier detection on each type of data in the sensor dataset using the absolute deviation of the local median in a sliding window, remove the data points marked as outliers in the sensor dataset, and fill the gaps using linear interpolation.

[0028] In this step, outlier detection is performed on various types of data in the sensor dataset using a sliding window local median absolute deviation (MAD). Data points marked as outliers are removed, and gaps are filled using linear interpolation. Specifically, taking a one-dimensional time series of any type of sensor data as an example, a sliding time window of length W is used. Within each window, the local median and the median of the absolute deviation relative to the median are calculated to obtain the local median absolute deviation (MAD) of that window. For the sampled value at the center of the window, the ratio of its absolute deviation relative to the local median to the MAD is calculated. If this ratio exceeds a preset outlier threshold, the sampled value is marked as an outlier data point. After processing the entire time series, all data points marked as outliers are removed from the sequence, and their corresponding positions are considered gaps. For each gap interval, linear interpolation between the nearest normal sampled points on both sides is used to fill the gap, thereby obtaining the sensor data sequence after local median absolute deviation detection and linear interpolation repair.

[0029] Step S13: After interpolating and filling the sensor dataset, calculate the mean of each type of data in the sensor dataset and perform mean removal processing to output BeiDou positioning coordinate data and laser displacement sensing data.

[0030] In this step, after interpolation and imputation of the sensor dataset, the mean of each data type in the sensor dataset is calculated and mean-removing processing is performed to output BeiDou positioning coordinate data and laser displacement sensing data. Specifically, after outlier removal and linear interpolation imputation, the arithmetic mean of each coordinate component of the BeiDou positioning coordinate data and the laser displacement sensing data is calculated over the entire observation period. Then, the sampled values ​​at each time point are subtracted from the corresponding mean to obtain the BeiDou positioning coordinate time series and laser displacement time series with a mean of zero. Mean-removing processing can eliminate static bias components, ensuring that the output BeiDou positioning coordinate data and laser displacement sensing data only reflect the dynamic vibration changes of the structure.

[0031] In this embodiment, coarse outlier detection based on solution quality indicators can prioritize the removal of data segments with obviously unreliable positioning solution quality, avoiding significant interference from low-quality solutions to subsequent processing. Fine detection using the absolute deviation of the local median in a sliding window can identify anomalous data such as abrupt changes, spikes, and short-term drifts at a local scale, thus cleaning up noise more accurately. Furthermore, linear interpolation is used to smoothly fill gaps, ensuring the continuity of the time series and avoiding time alignment deviations and feature distortions caused by data breakpoints. The mean of each type of data after interpolation is calculated and mean-removed, so that both BeiDou positioning coordinate data and laser displacement data are represented in zero-mean form, which is beneficial for highlighting dynamic change characteristics such as vibration.

[0032] Step S2: Interpolate the initial BeiDou positioning time series in the BeiDou positioning coordinate data and the initial laser displacement time series in the laser displacement sensing data to a common time reference to obtain a BeiDou positioning time series and a laser displacement time series located on a unified time axis.

[0033] In this step, the initial BeiDou positioning time series and the initial laser displacement time series are resampled onto a common time reference based on the piecewise cubic Hermite interpolation method, and the resampled initial BeiDou positioning time series are filtered to obtain the BeiDou positioning time series and laser displacement time series located on a unified time axis.

[0034] Furthermore, to avoid information loss, the higher value from the original sampling rate was selected as the benchmark to generate a common time series. Both datasets were interpolated using piecewise cubic Hermite interpolation.

[0035] h interp =PCHIP( t 1, h , t interp ) in, t1 represents the original time point of the initial BeiDou positioning time series or the initial laser displacement time series. h This indicates the magnitude of the sequence values ​​after removing the mean. t interp This represents the common time point of the initial BeiDou positioning time series and the initial laser displacement time series, generated according to the target sampling interval based on the start and end times.

[0036] After interpolation, a second-order Butterworth low-pass filter is applied to the BeiDou positioning time series to suppress high-frequency noise.

[0037] f cut = min (2.0, 0.9× f Nyquist ) in, H ( s ) represents the transfer function of the low-pass filter in the Laplace domain. s Represents the complex frequency variable in the Laplace transform domain. n This indicates the order of the low-pass filter. ω c This indicates the cutoff angular frequency of the low-pass filter. f cut Indicates the cutoff frequency. f Nyquist This represents the Nyquist frequency, which is half of the target frequency.

[0038] Then, zero-phase filtering is applied to the low-pass filtered BeiDou positioning time series:

[0039] in, h filtered This represents the filtered, unified time-axis signal, namely the BeiDou positioning time series and the laser displacement time series located on the unified time axis. H ( s ) represents the transfer function of the low-pass filter in the Laplace domain. H ( s -1 The ) denotes the inverse filtering operator corresponding to the forward and reverse filtering of a signal with a unified time axis when implementing zero-phase filtering. h interp This represents the initial BeiDou positioning time series mapped onto a unified time axis after resampling and interpolation.

[0040] In this embodiment, piecewise cubic Hermite interpolation is used to resample the initial time series of BeiDou and laser to the same common time reference, ensuring point-to-point correspondence between the two on a unified time axis and eliminating structural time errors caused by different sampling frequencies and inconsistent sampling times. Compared with simple linear interpolation, cubic Hermite interpolation has higher-order smoothness on the time series, better preserving the trend of the original signal while ensuring smooth interpolation. Piecewise cubic Hermite interpolation can utilize the derivative information of adjacent data points to more accurately reconstruct the signal shape within the interpolation segment, with fewer distortions such as overshoot and ringing, which is beneficial for preserving key dynamic features such as vibration and displacement changes. Compared with conventional cubic splines, Hermite interpolation is easier to control the local shape and can improve the stability and physical rationality of interpolation under conditions of local data abrupt changes or uneven sampling. Filtering the resampled BeiDou positioning time series can effectively suppress high-frequency noise, random jitter, and pseudo-high-frequency components introduced by interpolation, making the BeiDou trajectory smoother and closer to the real motion on a unified time axis.

[0041] Step S3: Perform Hilbert transform on the laser displacement time series to obtain the instantaneous envelope of the analytical signal corresponding to the laser displacement time series, and identify the vibration segment in the laser displacement time series based on the instantaneous envelope.

[0042] This step also includes steps S31 to S33.

[0043] Step S31: Perform Hilbert transform on the laser displacement time series to obtain the Hilbert transform components corresponding to the laser displacement time series, and construct an analytical signal based on the Hilbert transform components.

[0044] In this step, the laser displacement time series is represented by the Hilbert transform as follows:

[0045] in, h ( t This represents the laser displacement time series on a unified time axis, i.e., the laser sensor at time t. t Displacement signal; Representing the laser displacement time series h ( t The Hilbert transform of the original signal is the imaginary part of the analytic signal. pv This indicates that the integral is calculated according to the meaning of Cauchy's principal value, and is used to define the principal value integral of the Hilbert transform. τ Indicates time τ .

[0046] Step S32: Calculate the instantaneous envelope corresponding to the analytical signal, and smooth the instantaneous envelope by median filtering to obtain a smooth envelope.

[0047] In this step, the instantaneous envelope corresponding to the analytic signal is calculated as follows:

[0048]

[0049] The instantaneous envelope smoothed by median filtering is represented as: E smooth ( t =MedianFilter( E ( t ), W median ) in, z ( t This represents an analytic signal composed of a laser displacement time series and its Hilbert transform. j Represents the imaginary unit. E ( t () represents the instantaneous envelope corresponding to the analytic signal. E smooth ( t ) represents the instantaneous envelope E ( t The smooth envelope sequence obtained after median filtering; MedianFilter() represents the median filtering operator, which replaces the center point with the median of the samples in a given window; W median This represents the typical number of sampling points or window length for median filtering, i.e., the number of adjacent sampling points involved in a single median operation.

[0050] Step S33: Calculate the vibration detection threshold based on the smooth envelope, and mark the time interval in which the smooth envelope continuously exceeds the vibration detection threshold and the duration exceeds the preset minimum duration as a vibration segment.

[0051] In this step, the vibration detection threshold is calculated as follows: E th =median( E smooth )+ k ·std( E smooth ) in, E th Indicates the vibration detection threshold, median( E smooth) represents the median of the smooth envelope sequence, std( E smooth ) represents the standard deviation of the smooth envelope sequence. k This represents the sensitivity coefficient, which typically ranges from 0.5 to k and from 2.

[0052] Step S4: Calculate the time offset between the BeiDou positioning time series and the laser displacement time series according to the segmented cross-correlation algorithm, and calculate the confidence level corresponding to each vibration segment.

[0053] In this embodiment, for each vibration segment, the maximum search range is first set. L max Calculate the normalized cross-correlation:

[0054] in, R 12 ( τ This indicates that the BeiDou positioning time series and the laser displacement time series have a time lag. τ The normalized cross-correlation function value under the following conditions τ This represents the discrete-time lag. h 1 indicates a BeiDou time series that has been mean-removed and interpolated to a common time point. h 2 represents the laser displacement time series after mean removal and interpolation to the common time point.

[0055] The time offset can be obtained by finding the position with the largest absolute value of the correlation coefficient using the cross-correlation function. Δt and its confidence level C Represented as:

[0056]

[0057]

[0058] in, Δt This represents the time offset of the BeiDou positioning time series relative to the laser displacement time series within a certain vibration segment. τ* This represents the expression that makes the absolute value of the cross-correlation function | R 12 ( τ The optimal lag value corresponding to the attainment of the maximum value. Δt ds It represents the time interval between two adjacent sampling points on a unified time axis.

[0059] In this step, the time offset rationality check specifically includes: if the difference between the time offsets of any two adjacent vibration segments is greater than the preset continuity threshold, then the two adjacent vibration segments are marked as discontinuous abnormal segments; if the absolute value of the time offset of any vibration segment exceeds the preset time offset threshold, then the vibration segment is marked as an abnormal segment that exceeds the reasonable range; if the confidence level corresponding to any vibration segment is less than the preset confidence level threshold, then the vibration segment is marked as a low-quality abnormal segment.

[0060] Understandably, the confidence level is... C The segment with the greatest vibration was used as the reference segment, and its time offset was recorded. Δt ref As a global benchmark, if there is no high-confidence segment, use all C The median offset is greater than 0.2. Outlier detection is performed on each offset segment.

[0061] (1) Continuity check: If the offset difference between adjacent segments is too large, it is marked as discontinuous; (2) Physical feasibility check: If the offset is unreasonable, mark it as exceeding the reasonable range; (3) Low confidence check: If C A value <0.3 is marked as low quality. For aberrant segments, different corrections are applied based on their location. An aberrant segment at the beginning is directly corrected to [value missing]. Δt ref The anomaly at the end is corrected to the offset of the preceding segment, and the anomaly in the middle segment is corrected to (the offset of the preceding segment + ...). Δt ref ) / 2.

[0062] Furthermore, the incremental correction operation specifically includes: Select a reference vibration segment from the vibration segments with the highest confidence level, and use the time offset corresponding to the reference vibration segment as the global time offset reference. For abnormal vibration segments located at the beginning of the BeiDou positioning time series or laser displacement time series, the time offset of the abnormal vibration segments is directly corrected to the global time offset reference. For abnormal vibration segments located at the end of the BeiDou positioning time series or laser displacement time series, the time offset of the abnormal vibration segment is corrected to the time offset of the previous non-abnormal vibration segment. For abnormal vibration segments located in the middle of the BeiDou positioning time series or laser displacement time series, the time offset of the abnormal vibration segment is corrected to the average of the time offset of the previous non-abnormal vibration segment and the global time offset benchmark.

[0063] In this embodiment, the vibration segment with the highest confidence level is used as the reference vibration segment, and its corresponding time offset is used as the global time offset reference, so that the overall correction result is anchored on the most reliable data segment. This avoids local outlier segments or low-confidence segments dominating the global time offset, ensuring a high level of reliability for the global time reference. For outlier vibration segments at the beginning of the sequence, their time offsets are directly corrected to the global time offset reference, preventing the amplified impact on the overall time series when the initial data lacks reference and has large errors. For outlier vibration segments at the end of the sequence, their time offsets are corrected to the time offset of the preceding non-outlier segment, avoiding time offset jumps at the end due to insufficient data or high noise. By performing differentiated processing on outlier vibration segments based on their location characteristics in the time series, different correction strategies are adopted to specifically reduce the impact of local anomalies. Even if multiple consecutive or scattered outlier segments exist, error propagation can be limited under the constraint of the preceding non-outlier segment plus the global reference, preventing the global time offset estimate from being dragged away from the true value.

[0064] like Figure 2 As shown, Figure 2 A comparison of the time-aligned BeiDou and laser displacement responses is shown. A three-axis linear module vibration table was used, with the BeiDou monitoring receiver and laser displacement sensor mounted on it for vibration data acquisition. Because the laser sensor uses a sliding window averaging method to smooth displacement measurements to ensure accuracy, there is still an output delay even when both sensors use the same hardware clock. Therefore, this method can be used for time alignment. The vibration mode was set to a vertical sine wave with an amplitude of 16mm and a frequency of 0.75Hz. A schematic diagram of the vibration time series is shown below. Figure 2 As shown, a total of 15 sets of vibrations were continuously performed. In the intervals of approximately 690–700 s and 750–760 s, the displacement amplitude of the curve was small, with only slight fluctuations, indicating that the structure was in a non-vibration or weak vibration state. In the interval of approximately 700–750 s, the curve exhibited strong periodic vibration, with the amplitude varying between ±10 millimeters. The waveforms of the two sensors almost completely overlapped, indicating that their measurement results during this period were highly consistent.

[0065] The time alignment accuracy evaluation index is represented by the standard deviation of the difference between the mean-reduced BeiDou coordinate time series and the laser displacement time series, i.e.

[0066] in, h 1 indicates a BeiDou time series that has been mean-removed and interpolated to a common time point. h 2 represents the laser displacement time series interpolated to a common time point after removing the mean. This example compares and analyzes the standard deviation obtained by calculating the time offset without time alignment, calculating the entire time offset, and calculating the time offset using this method.

[0067] Step S5: Based on the time offset and confidence level of all vibration segments, perform time offset rationality verification and progressive correction operations to obtain global time offset, and globally correct and output the timestamps of Beidou positioning time series or laser displacement time series according to the global time offset.

[0068] In this step, the time offsets of all vibration segments are interpolated or extrapolated on the time axis to obtain continuous time offset curves, and all timestamps of the BeiDou positioning time series or laser displacement time series are globally corrected based on the time offset curves.

[0069] Further, the time offsets of each vibration segment obtained in step S4 and their corresponding confidence levels are combined to form a time offset set. Vibration segments with confidence levels below a preset lower limit are removed, and the median of the time offsets and the median of their absolute deviations are calculated for the remaining vibration segments. When the deviation of a vibration segment's time offset exceeds a preset multiple threshold, the time offset of that vibration segment is marked as unreasonable and removed, thus obtaining a set of vibration segments that pass the time offset reasonableness test. Then, using the confidence level of each vibration segment as a weight, a weighted average or progressive recursion is performed on the time offsets that pass the test to obtain the global time offset. Based on this, using the unified time axis of the laser displacement time series as a reference, the timestamps of each epoch of the BeiDou positioning time series are globally corrected to obtain the corrected timestamps; or conversely, the timestamps of the laser displacement time series are globally shifted and corrected.

[0070] like Figure 3 The figures show the time series plots of BeiDou coordinates versus laser displacement residuals, obtained without time alignment, with the entire time offset calculated, and with the time offset calculated using the proposed method. It can be seen that the displacement residuals obtained by the aligned method and the proposed method are closer to zero compared to the unaligned method.

[0071] like Figure 4 As shown, Figure 4 (a) in the figure is the normal distribution of error under the misalignment method. Figure 4 (b) in the figure is the error normal distribution plot under the whole segment alignment method. Figure 4 (c) in the figure is the error normal distribution diagram of this scheme. In order to more clearly compare the time alignment accuracy difference between the whole segment alignment method and this method, the normal distribution results of the three residual time series are compared. Figure 4 The results show that the standard deviation of the displacement difference obtained by this method is 4.86 mm, which is better than both the misalignment method and the whole segment alignment method.

[0072] In this embodiment, by uniformly interpolating the BeiDou positioning time series and the laser displacement time series to a common time reference, and combining piecewise cross-correlation and vibration feature identification, precise time synchronization of the two types of heterogeneous sensor data is achieved. This significantly reduces time deviation and improves the accuracy of multi-source data fusion. Furthermore, by obtaining the instantaneous envelope of the laser displacement signal through Hilbert transform and automatically identifying vibration segments, time offset estimation is performed only in time periods containing obvious dynamic characteristics. This effectively avoids invalid or unreliable alignment calculations during static or low-dynamic phases. Simultaneously, a piecewise cross-correlation algorithm is used to calculate the time offset of each vibration segment, and a confidence index is assigned to each offset. This effectively quantifies the contribution and reliability of different time periods to the alignment result. Using the time offsets and confidence scores obtained from all vibration segments, the rationality of the time offset is checked, eliminating or weakening offset results inconsistent with the overall trend. A global time offset is obtained through progressive correction, realizing the transition from local estimation to a globally unified offset, reducing the impact of single local segment errors on the final result, and improving the accuracy and reliability of time alignment. Based on the above method, this application discloses an adaptive time alignment system for BeiDou and laser sensors based on multi-source data fusion, with reference to... Figure 5 The BeiDou and laser sensor adaptive time alignment system 1 includes a data acquisition module 11, a data processing module 12, and a global correction module 13, wherein... The data acquisition module 11 is used to acquire the sensing datasets of multiple modal sensors and preprocess the sensing datasets to obtain BeiDou positioning coordinate data and laser displacement sensing data. The data processing module 12 is used to interpolate the initial BeiDou positioning time series in the BeiDou positioning coordinate data and the initial laser displacement time series in the laser displacement sensing data to a common time reference to obtain the BeiDou positioning time series and laser displacement time series located on a unified time axis. The module performs Hilbert transform on the laser displacement time series to obtain the instantaneous envelope of the analytical signal corresponding to the laser displacement time series. Based on the instantaneous envelope, the module identifies the vibration segments in the laser displacement time series. The module calculates the time offset between the BeiDou positioning time series and the laser displacement time series according to the piecewise cross-correlation algorithm and calculates the confidence level corresponding to each vibration segment. The global correction module 13 is used to perform time offset rationality verification and progressive correction operations based on the time offset and confidence of all vibration segments to obtain the global time offset, and to globally correct and output the timestamps of the Beidou positioning time series or laser displacement time series based on the global time offset.

[0073] In one example, the data acquisition module 11 is used to perform coarse outlier detection on the sensor dataset based on the solution quality identifier, which includes fixed solution or floating-point solution; fine outlier detection is performed on each type of data in the sensor dataset using the absolute deviation of the local median of a sliding window; data points marked as outliers in the sensor dataset are removed and gaps are filled by linear interpolation; after the interpolation filling of the sensor dataset is completed, the mean of each type of data in the sensor dataset is calculated and the mean is removed to output BeiDou positioning coordinate data and laser displacement sensing data.

[0074] In one example, the data processing module 12 is used to resample the initial BeiDou positioning time series and the initial laser displacement time series onto a common time reference based on the piecewise cubic Hermite interpolation method, and to filter the resampled initial BeiDou positioning time series to obtain the BeiDou positioning time series and laser displacement time series located on a unified time axis.

[0075] In one example, the data processing module 12 is used to perform Hilbert transform on the laser displacement time series to obtain the Hilbert transform components corresponding to the laser displacement time series, and construct an analytical signal based on the Hilbert transform components; calculate the instantaneous envelope corresponding to the analytical signal, and smooth the instantaneous envelope by median filtering to obtain a smooth envelope; calculate the vibration detection threshold based on the smooth envelope, and mark the time interval in which the smooth envelope continuously exceeds the vibration detection threshold and the duration exceeds the preset minimum duration as a vibration segment.

[0076] In one example, the time offset plausibility check operation specifically includes: If the difference in time offset between any two adjacent vibration segments is greater than the preset continuity threshold, then the two adjacent vibration segments are marked as discontinuous abnormal segments. If the absolute value of the time offset of any vibration segment exceeds the preset time offset threshold, the vibration segment will be marked as an abnormal segment that exceeds the reasonable range. If the confidence level of any vibration segment is less than the preset confidence threshold, the vibration segment will be marked as a low-quality abnormal segment.

[0077] In one example, the incremental correction operation specifically includes: Select a reference vibration segment from the vibration segments with the highest confidence level, and use the time offset corresponding to the reference vibration segment as the global time offset reference. For abnormal vibration segments located at the beginning of the BeiDou positioning time series or laser displacement time series, the time offset of the abnormal vibration segments is directly corrected to the global time offset reference. For abnormal vibration segments located at the end of the BeiDou positioning time series or laser displacement time series, the time offset of the abnormal vibration segment is corrected to the time offset of the previous non-abnormal vibration segment. For abnormal vibration segments located in the middle of the BeiDou positioning time series or laser displacement time series, the time offset of the abnormal vibration segment is corrected to the average of the time offset of the previous non-abnormal vibration segment and the global time offset benchmark.

[0078] In one example, the global correction module 13 is used to interpolate or extrapolate the time offset of all vibration segments on the time axis to obtain a continuous time offset curve, and to globally correct all timestamps of the BeiDou positioning time series or laser displacement time series based on the time offset curve.

[0079] Please see Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Figure 6 As shown, the electronic device 2 may include: at least one processor 21, at least one network interface 24, user interface 23, memory 25, and at least one communication bus 22.

[0080] The communication bus 22 is used to enable communication between these components.

[0081] The user interface 23 may include a display screen and a camera. Optionally, the user interface 23 may also include a standard wired interface and a wireless interface.

[0082] The network interface 24 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0083] The processor 21 may include one or more processing cores. The processor 21 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in the memory 25, and by calling data stored in the memory 25. Optionally, the processor 21 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 21 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also be implemented as a separate chip without being integrated into the processor 21.

[0084] The memory 25 may include random access memory (RAM) or read-only memory. Optionally, the memory 25 may include non-transitory computer-readable storage medium. The memory 25 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 25 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch function, sound playback function, image playback function, etc.), instructions for implementing the above-described method embodiments, etc.; the data storage area may store data involved in the above-described method embodiments, etc. Optionally, the memory 25 may also be at least one storage device located remotely from the aforementioned processor 21. Figure 6 As shown, the memory 25, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for an adaptive time alignment method for BeiDou and laser sensors based on vibration characteristics.

[0085] exist Figure 6In the electronic device 2 shown, the user interface 23 is mainly used to provide an input interface for the user and to obtain the user input data; while the processor 21 can be used to call an application program stored in the memory 25 that is a Beidou and laser sensor adaptive time alignment method based on vibration characteristics. When executed by one or more processors, the electronic device performs one or more methods as described in the above embodiments.

[0086] A computer-readable storage medium storing instructions that, when executed by one or more processors, cause a computer to perform one or more methods as described in the embodiments above.

[0087] It should be noted that, for the sake of simplicity, the foregoing method embodiments are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, as some steps may be performed in other orders or simultaneously according to this application. Furthermore, those skilled in the art should also understand that the embodiments described in the specification are preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0088] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0089] In the several embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the shown or discussed mutual couplings or direct couplings or communication connections may be through some service interfaces; indirect couplings or communication connections between apparatuses or units may be electrical or other forms.

[0090] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0091] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0092] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0093] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for adaptive time alignment between BeiDou and laser sensors based on vibration characteristics, characterized in that, The method includes: Collect sensor datasets from multiple modal sensors and preprocess the sensor datasets to obtain BeiDou positioning coordinate data and laser displacement sensing data. The initial BeiDou positioning time series in the BeiDou positioning coordinate data and the initial laser displacement time series in the laser displacement sensing data are interpolated to a common time reference to obtain a BeiDou positioning time series and a laser displacement time series located on a unified time axis. Perform Hilbert transform on the laser displacement time series to obtain the instantaneous envelope of the analytical signal corresponding to the laser displacement time series, and identify the vibration segment in the laser displacement time series based on the instantaneous envelope; The time offset between the BeiDou positioning time series and the laser displacement time series is calculated using a segmented cross-correlation algorithm, and the confidence level corresponding to each vibration segment is calculated. Based on the time offset and confidence level of all vibration segments, a time offset rationality check and progressive correction operation are performed to obtain a global time offset. Then, the timestamps of the BeiDou positioning time series or laser displacement time series are globally corrected and output according to the global time offset.

2. The adaptive time alignment method for BeiDou and laser sensors based on vibration characteristics as described in claim 1, characterized in that, The preprocessing of the sensor dataset specifically includes: The sensor dataset is subjected to coarse outlier detection based on the solution quality identifier, wherein the solution quality identifier includes fixed solution or floating-point solution; Outlier detection is performed on each type of data in the sensor dataset using a sliding window local median absolute deviation. Data points marked as outliers in the sensor dataset are removed and gaps are filled using linear interpolation. After interpolation and imputation of the sensor dataset, the mean of each type of data in the sensor dataset is calculated and the mean is removed to output the BeiDou positioning coordinate data and the laser displacement sensing data.

3. The adaptive time alignment method for BeiDou and laser sensors based on vibration characteristics as described in claim 1, characterized in that, The acquisition of the BeiDou positioning time series and laser displacement time series located on a unified time axis specifically includes: The initial BeiDou positioning time series and the initial laser displacement time series are resampled onto a common time reference using the piecewise cubic Hermite interpolation method, and the resampled initial BeiDou positioning time series are filtered to obtain a BeiDou positioning time series and a laser displacement time series located on a unified time axis.

4. The adaptive time alignment method for BeiDou and laser sensors based on vibration characteristics as described in claim 1, characterized in that, The step of identifying vibration segments in the laser displacement time series based on the instantaneous envelope specifically includes: Perform Hilbert transform on the laser displacement time series to obtain the Hilbert transform components corresponding to the laser displacement time series, and construct an analytical signal based on the Hilbert transform components; Calculate the instantaneous envelope corresponding to the analytic signal, and smooth the instantaneous envelope by median filtering to obtain a smooth envelope; The vibration detection threshold is calculated based on the smooth envelope, and the time interval in which the smooth envelope continuously exceeds the vibration detection threshold and the duration exceeds the preset minimum duration is marked as a vibration segment.

5. The adaptive time alignment method for BeiDou and laser sensors based on vibration characteristics as described in claim 1, characterized in that, The time offset reasonableness check operation specifically includes: If the difference in time offset between any two adjacent vibration segments is greater than a preset continuity threshold, then the two adjacent vibration segments are marked as discontinuous abnormal segments. If the absolute value of the time offset of any vibration segment exceeds the preset time offset threshold, the vibration segment will be marked as an abnormal segment that is outside the reasonable range. If the confidence level corresponding to any vibration segment is less than the preset confidence threshold, then the vibration segment is marked as a low-quality abnormal segment.

6. The adaptive time alignment method for BeiDou and laser sensors based on vibration characteristics as described in claim 5, characterized in that, The gradual correction operation specifically includes: Select a reference vibration segment from the vibration segments with the highest confidence level, and use the time offset corresponding to the reference vibration segment as the global time offset reference. For abnormal vibration segments located at the beginning of the BeiDou positioning time series or the laser displacement time series, the time offset of the abnormal vibration segments is directly corrected to the global time offset reference. For abnormal vibration segments located at the end of the BeiDou positioning time series or the laser displacement time series, the time offset of the abnormal vibration segment is corrected to the time offset of the previous non-abnormal vibration segment. For abnormal vibration segments located in the middle of the BeiDou positioning time series or the laser displacement time series, the time offset of the abnormal vibration segment is corrected to the average of the time offset of the previous non-abnormal vibration segment and the global time offset reference.

7. The adaptive time alignment method for BeiDou and laser sensors based on vibration characteristics as described in claim 1, characterized in that, The step of globally correcting and outputting the timestamps of the BeiDou positioning time series or laser displacement time series based on the global time offset specifically includes: Interpolate or extrapolate the time offsets of all vibration segments on the time axis to obtain a continuous time offset curve, and perform global correction on all timestamps of the BeiDou positioning time series or the laser displacement time series based on the time offset curve.

8. A BeiDou and laser sensor adaptive time alignment system based on vibration characteristics, characterized in that, The BeiDou and laser sensor adaptive time alignment system (1) includes a data acquisition module (11), a data processing module (12), and a global correction module (13), wherein, The data acquisition module (11) is used to acquire the sensing datasets of multiple modal sensors and preprocess the sensing datasets to obtain BeiDou positioning coordinate data and laser displacement sensing data. The data processing module (12) is used to interpolate the initial BeiDou positioning time series in the BeiDou positioning coordinate data and the initial laser displacement time series in the laser displacement sensing data to a common time reference to obtain a BeiDou positioning time series and a laser displacement time series located on a unified time axis. The module performs Hilbert transform on the laser displacement time series to obtain the instantaneous envelope of the analytical signal corresponding to the laser displacement time series. Based on the instantaneous envelope, the module identifies the vibration segments in the laser displacement time series. The module calculates the time offset between the BeiDou positioning time series and the laser displacement time series according to the segmented cross-correlation algorithm and calculates the confidence level corresponding to each vibration segment. The global correction module (13) is used to perform time offset rationality verification and progressive correction operations based on the time offset and confidence of all vibration segments to obtain the global time offset, and to perform global correction and output of the timestamp of the Beidou positioning time series or laser displacement time series according to the global time offset.

9. An electronic device, characterized in that, The device includes a processor (21), a memory (25), a user interface (23), and a network interface (24). The memory (25) is used to store instructions. The user interface (23) and the network interface (24) are used to communicate with other devices. The processor (21) is used to execute the instructions stored in the memory (25) to cause the electronic device (2) to perform the method as described in any one of claims 1-7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1-7.

Images